This is a mirror of https://gitlab.com/cinematheque-quebecoise/cq2rdf

All contributions must be done through Gitlab.

cq2rdf

A tool to convert the database of Cinemathèque québécoise in multple RDF formats.

Read this in other languages: English, Français.

Prerequisites

1. SQLite database file containing Cinemathèque québécoise's data (ex. cinetv-2019-07-12-publique.db). See releases from repository cinetvdb.
2. cabal or Nix must be installed.

Usage

$ cq2rdf-exe --help
cq2rdf Copyright (C) 2020 Cinemathèque québécoise This program comes with
ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to
redistribute it under certain conditions.

Usage: cq2rdf-exe [--version] [--help] [-v|--verbose] COMMAND
                  (-b|--baseuri BASEURI) (-s|--sqlitedb SQLITEDBFILE)
                  (-o|--outputdir OUTPUTDIR)
  cq2rdf is a tool to convert the database of Cinemathèque québécoise in multple
  RDF formats.

Available options:
  --version                Show version
  --help                   Show this help text
  -v,--verbose             Verbose output?
  -b,--baseuri BASEURI     RDF Base URI
  -s,--sqlitedb SQLITEDBFILE
                           File path of the Sqlite database file
  -o,--outputdir OUTPUTDIR Output directory of result

Available commands:
  cinetv-to-rdf            Convert CineTV to RDF
  adlib-to-rdf             Convert Adlib to RDF (not yet implemented)
  generate-void            Generate a VoID dataset from SPARQL endpoint

Environnement de développement avec Nix

With Nix installed, go inside a dev environment with:

$ nix-shell

Development

To compile the program:

$ cabal build

It can be executed with:

$ cabal run cq2rdf-exe -- --helpgitlab mirror different README

Predefined commands are available in the Makefile such as:

$ make build
$ make cinetv-to-rdf
$ make generate-void

Generating executable with Nix

Generate an executable file with:

$ make build

Which will be available in ./result/bin/cq2rdf-exe.

Unit tests

Unit tests can be launched with cabal test.

To configure Cabal with HPC for code coverage:

cabal clean
cabal configure --enable-tests --enable-coverage

Then, you can simply execute unit tests with cabal test.

SPARQL queries tests

Predefined SPARQL queries are available in the file example-queries.yaml. They are used to valide the correctness of the generated RDF files. Blazegraph is used to test those queries.

First off, execute make bootstrap-blazegraph to import the RDF data inside Blazegraph. Afterwards, execute make test-sparql-queries to test the SPARQL queries.

Dataset description with VoID

The generated data can be described using the VoID ontology.

Some default information on the RDF dataset is defined in the file void.base.ttl. These values can be changed if applicable.

Other statements, such as number of triples with void:triples, are automatically inserted in the resulting file void.ttl.

Data publication

Be sure to upgrade the version inside package.yaml.

Before publishing the RDF data, you need to generate a token from Gitlab's web interface. This token is called a Personal Access Token. Choose a name, an expiration date and the "api" scope. Put the generated token in the hidden local file .gitlab-token.

To publish the data in the folder cmtq-dataset, simply run:

$ make release

License

cq2rdf is licenced in GPLv3. See LICENSE for a complete copy of the LICENSE.